Qinglan Ding, PhD, on Associations Between Polysomnographic Phenotypes of OSA and Incident Type 2 Diabetes
In this podcast, Qinglan "Priscilla" Ding, PhD, talks about her research team's study on the associations between obstructive sleep apnea (OSA) and incident type 2 diabetes, including the relevance of the 7 polysomnographic phenotypes for risk of type 2 diabetes in patients who have undergone OSA evaluation.
- Muraki I, Wada H, Tanigawa T. Sleep apnea and type 2 diabetes. J Diabetes Investig. 2018;9(5):991-997. https://doi.org/10.1111/jdi.12823
- Zinchuk AV, Jeon S, Koo BB, et al. Polysomnographic phenotypes and their cardiovascular implications in obstructive sleep apnoea. Thorax. 2018;73(5):472-480. https://doi.org/10.1136/thoraxjnl-2017-210431
- Zinchuk AV, Yaggi HK. Phenotypic subtypes of OSA: a challenge and opportunity for precision medicine. Chest. 2020;157(2):403-420. https://doi.org/10.1016/j.chest.2019.09.002
- Zinchuk, AV, Gentry MJ, Concato J, Yaggi HK. Phenotypes in obstructive sleep apnea: a definition, examples, and evolution of approaches. Sleep Med Rev. 2017;35:113-123. https://doi.org/10.1016/j.smrv.2016.10.002
Qinglan "Priscilla" Ding, PhD, is a health outcomes researcher and an assistant professor at the School of Nursing, College of Health and Human Science, and a faculty associate at the Center on Aging and the Life Course at Purdue University in West Lafayette, Indiana.
Published in partnership with The American Thoracic Society.
Jessica Bard: Hello everyone and welcome to another installment of "Podcast 360," your go‑to resource for medical news and clinical updates. I'm your moderator Jessica Bard with Consultant360 Specialty Network.
Type 2 diabetes is a major public health concern with high morbidity, mortality and health care costs. Research indicates the majority of patients with type 2 diabetes also have obstructive sleep apnea.
Dr Priscilla Ding is here to speak with us about that today. Dr Ding is a health outcomes researcher at Purdue University. Thank you for joining us today, Dr Ding.
You're presenting a session titled, "Associations Between Polysomnographic Phenotypes of Obstructive Sleep Apnea and Incident Type 2 Diabetes at HCS 2021." Can you please give us an overview of your session?
Dr Priscilla Ding: It's my privilege to be here today. I had the privilege to collaborate with Dr Andrey V Zinchuk and Dr Henry K Yaggi, both of whom from Yale University in the study. Their research team had previously identified seven obstructive sleep apnea phenotypes from different polysomnography matrix, using unsupervised machine learning methods.
These phenotypes were named according to their distinguishing polysomnography features. For example, one of the seven phenotypes was named Periodic Limb Movement of Sleep. That's because it was characterized by having the highest periodic limb movement of sleep index and a low respiratory event frequency as indicated by relevant polysomnography metrics.
In addition, Dr Zinchuk and Yaggi's research team previously discovered that objective sleep apnea patients with different polysomnographic phenotypes had different levels of risk for cardiovascular disease and death during a five‑year follow up, because our common interest in the associations between cardio‑metabolic risk and sleep disorders and because we suspected that the pathophysiological process in the causal pathway between OSA and cardiovascular disease, may also link OSA with type 2 diabetes.
We conducted the study to determine the relevance of the seven polysomnographic phenotypes for risk of type 2 diabetes in patients who went through OSA evaluation.
In this study, we used the data from the Determining Risk of Vascular Events by Apnea Monitoring, abbreviated as the DREAM study, which was a clinic‑based observational study of a cohort of veterans referred for all evaluations at veteran centers in Connecticut, Indiana and Ohio. None of the patients had type 2 diabetes at baseline enrollment.
In our study, in addition to determine the relationship between the seven polysomnographic phenotypes and newly developed type 2 diabetes during follow up among these veterans, we also compare the predictive value of polysomnographic phenotype for type 2 diabetes with that of the apnea‑hypopnea index, commonly abbreviated as AHI.
Currently, AHI is the metric used to describe the severity of OSA. AHI measure the intermittent hypoxia and arousals during sleep in patients with OSA. In my section, we'll be covering in more details about the study methods and our main findings resulting from the study analysis.
Jessica: You mentioned some comorbidities and associations, but if you could dig in a little bit deeper. What are the comorbidities, and also the associations of the OSA in incident type 2 diabetes?
Dr Ding: In terms of the comorbidities of OSA in incident type 2 diabetes, OSA and type 2 Diabetes are both well‑known risk factors for cardiovascular disease. As we all know, cardiovascular disease is still the leading cause of death for both women and men in the United States, and in the world.
There is also increasing evidence suggesting that patients with OSA are more likely to develop type 2 diabetes, compared to those who didn't have OSA. Our study will be one of them adding to the current evidence. Also, I wanted to provide a little background about obstructive sleep apnea.
Obstructive sleep apnea ‑‑ which is commonly abbreviated as OSA ‑‑ is a potentially serious sleep disorder, in which breathing stop involuntarily for a brief period during sleep. It's seen in all age group, but it's more common in older adults and those who are obese. Normally, when we sleep, air flows smoothly from mouth and nose into the lungs at all times.
Periods when breathing stop are called apnea. In patients with OSA, the normal flow of air is repeatedly stopped throughout the night. The flow of air stop because the airway space in the area of the throat becomes too narrow. Snoring is characteristic of obstructive sleep apnea.
Untreated sleep apnea can cause health problems, especially different types of cardiovascular conditions, such hypertension, heart attack, and stroke. Recent studies also have found that patients with OSA have an increased risk for the development of diabetes.
Although, many studies in animals and humans have been conducted trying to explain the mechanisms link OSA with diabetes, we still don't know whether the associations between OSA and diabetes is due to the effect on insulin secretion, or is due to OSA being a marker of obesity and insulin resistance that increase the risk of type 2 diabetes, or is due to a shared mechanism. We just don't know the answers yet.
Jessica: Can you tell us more about what kinds of patients were involved in this study? Did any specific patient characteristics contribute to your results?
Dr Ding: In this study, we had 840 veterans referred for OSA evaluations at the three US Veterans Centers participated in our study. None of the veterans had base‑line type 2 diabetes when enrolled. The main age of these patients was 57 years.
They were primarily males with a medium body, mass index of 33.4, and a medium apnea hypopnea index of 12, which suggest most of them have mild OSA because an apnea‑hypopnea index range between 5 to 15, indicate the presence of mild OSA.
In terms of patient characteristics, contributing to our findings, as I said, our sample consisted of mainly male veterans and most of our participants were obese. As we all know, obesity is a risk factor for the development of type 2 diabetes and cardiovascular disease. That could potentially contribute to a higher risk of type 2 diabetes in the cohort.
However, we did additional analysis to adjust body mass index, the changes of weight over time, and the changes of BMI from baseline when assessing the associations between polysomnographic phenotypes, and insulin type 2 diabetes. The adjustment of those did not change the associations between polysomnographic phenotypes and type 2 diabetes.
Another factor to consider is that we have a primary male cohort, which could limit our ability to generalize our findings to the general population. Based on the results from some previous studies, gender did not modify the association between OSA and the risk of developing type 2 diabetes.
Jessica: How will this research impact clinical practice in the management of OSA and incident type 2 diabetes?
Dr Ding: Thanks for the question. In this study, we found that the incidence of new onset type 2 diabetes varied according to polysomnographic phenotypes, with the highest rate in the hypopnea and hypoxia phenotype, followed by the arousal and poor sleep phenotype, the combined severe phenotype, the rapid eye movement in hypoxia phenotype, and then the periodic limb movement of sleep phenotype, the mouth phenotype, and lastly, the non rapid eye movement and poor sleep phenotype.
Another finding is that we didn't find a linear relationship between apnea‑hypopnea index and type 2 diabetes incidence. For example, the combined severe phenotype had a low rate of incidence of type 2 diabetes during follow‑up, then the hypopnea and hypoxia group, but the hypopnea and hypoxia group had a low HI compared to the HI of the combined severe phenotype.
We also found that the hazard ratios of developing Incident type 2 diabetes during follow‑up was greatly increased for patients with hypopnea and hypoxia, and periodic limb movement of sleep phenotypes, as compared with those with a model phenotype for Incident type 2 diabetes.
When comparing type 2 diabetes predictive value between polysomnographic phenotype and HI, we found that adding polysomnographic phenotypes to the current known diabetes risk factors increase the predictive ability of future development of type 2 diabetes. HI did not, according to the findings of the likelihood ratio test in our analysis.
Going back to your question, in addition to further showing the link between OSA and incident type 2 diabetes, our study provide evidence that polysomnographic phenotypes can help predict type 2 diabetes among individuals who were referred for OSA evaluation.
We also realized that further studies are needed to confirm our findings in other cohorts, such as a cohort with more women, or a cohort from the general population before making any recommendations for clinicians to make changes in their practice.
However, our findings suggest that some high risk OSA subgroups might not be able to be identified by using HI alone.
Jessica: What would you say is next for research on this topic?
Dr Ding: As a next step, we're hoping to validate the seven polysomnographic phenotypes that will identify in the current study in external cohorts that have all the polysomnography metrics we used in this study. We're working hard to find such data sets and we welcome suggestions from our audience.
Additionally, we will include some of the newer metrics of physiological sleep disturbances, such as hypoxic burden and identifying polysomnographic phenotypes in future studies.
We plan to conduct a future study to evaluate whether treating OSA can change the associations between the different polysomnographic phenotypes and the risk of developing type 2 diabetes.
Jessica: Can you summarize what's the overall take home message for our audience?
Dr Ding: For the take home message is, first, I believe is essential for our audience to realize that there are independent associations between obstructive sleep apnea and cardio‑metabolic diseases such as heart disease and type 2 diabetes.
Secondly, they need to be aware that other polysomnographic metrics, in addition to the apnea‑hypopnea Index can be helpful to them in identifying high risk OSA patients, especially those with higher cardiovascular disease and diabetes risks.
Jessica: Is there anything else that you'd like to add? Anything that you think that we missed?
Dr Ding: Yes. Our study demonstrated the value of a research project involving a multidisciplinary research team. Our team have physicians and researchers from sleep medicine, endocrinology, neurology, pulmonology, statistics, nursing and engineering.
Each team member contributed to the development and implementation of the study, as well as the preparation of the findings reports. We are grateful for each of their unique contributions, without which we couldn't have been able to successfully complete our tasks.
We look forward to future collaborations together to discover and present more answers to the clinical and scientific community at large. I want to thank you all for listening and also thank you for the opportunity to share our study with all of you.
Jessica: Thank you for being with us today. We really appreciate your time and all of your work on this.
Dr Ding: Thank you.